18-899-K3-Carnegie Mellon University Africa - Carnegie Mellon University


Special Topics in Signal Processing: Data Analytics

Course discipline: Electrical and Computer Engineering
Core/Elective:  Core
Units: 6
Lecture/Lab/Rep hours/week: 
3 Lecture Hours Per Week
3 Lab Hours Per Week
Semester/year offered (fall/spring, even/odd/all years): Spring, all years
Pre-requisites:  Data and Inference and Applied Machine Learning Mini-Courses; Background in quantitative discipline (Engineering, Computer Science, Physics, Mathematics, Statistics); Programming.

Course description: 

This course will take a practical approach to solving challenges in the public and private sectors using data analytics.  A number of different themes will be explored as case studies in order to demonstrate how data-driven decision-making has widespread applications.  The course will examine how the question being posed, the available data and the selected modelling approach all come together to arrive at a feasible solution. A range of quantitative techniques, involving both linear and nonlinear methods will be presented for dealing with numerical structured datasets. Substantial emphasis will be placed on the process of delivering data analytics via a dashboard to facilitate decision-making and policy-making.  The course content will be structured to provide a roadmap for carrying out the necessary procedures and will be illustrated using case studies, reading material and previously published models. Participants will obtain hands-on experience by working on specific challenges with real-world data through a carefully structured set of assignments.  

Learning objectives:

The objective of this course is to provide students with practical experience of undertaking a data analytics project to address a challenge of their choice. This process involves: (1) proposing a feasible research plan based on a challenge that can be addressed with available data; (2) exploratory analysis of the data; (3) construction and evaluation of a quantitative model; and (4) communication of the model output to decision-makers.  This will involve identifying a relevant key performance indicator and collecting appropriate explanatory variables. The ability to write a report that documents the challenge, data, analysis, model, evaluation and feasibility of deploying the analytics in a real-world setting will be assessed. Applications will include forecasting, classification, novelty detection and risk monitoring.  


After completing this course, students should be able to: 
  • Design a data analytics project in response to a specific challenge
  • Download and organize data for addressing the challenge
  • Explore the dataset using visualization techniques
  • Apply a range of quantitative techniques
  • Discuss the advantages and disadvantages of different models
  • Select an approach that is optimal for meeting the objective
  • Present conclusions and recommendations
  • Communicate model output to decision-maker

Content details:

1. Weather

2. Renewable energy

3. Demand

4. Risk Assessment

5. Health and telemedicine

6. Early warning systems

Faculty: Patrick McSharry